An analysis of binary isotonic regression: degrees of freedom and implications for calibration

arXiv:2607.27301 · stat.ML, cs.LG, stat.ME · Submitted 2026-07-29 · Read on arXiv

Raphael Rossellini, Rina Foygel Barber, Zhimei Ren, Jake A. Soloff

stat.ML, cs.LG, stat.ME

Submitted: 2026-07-29

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Calibration — 38× in this paper · explained in 9 episodes
• Expected Calibration Error — 6× in this paper
• Consistency — 1× in this paper · explained in 3 episodes
• Exchangeability — 1× in this paper · explained in 2 episodes
• Monotonicity — 1× in this paper · explained in 2 episodes
• Regret — 1× in this paper · explained in 4 episodes

Sources

• In-sample calibration yields conformal calibration guarantees
• Can a calibration metric be both testable and actionable?

Related papers

• Behavior of prediction performance metrics with rare events
• Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
• A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
• Scalable Posterior Uncertainty for Flexible Density-Based Clustering
• A Transport Geometry of Benign Overfitting
• Efficient quadratic entropy with distance sketches